一、前言
tomcat具有某些特性,会对/;/
、/./
、/../
、/..;/
进行特殊处理,究竟是怎样的机制导致了这些特性,本文通过动态调试tomcat源码来分析url解析过程,解开疑惑
二、环境搭建
- jdk 8
- maven 3.6
- idea 2020.2
- tomcat 9.0.45
2.1 tomcat
从官网下载源码
复制conf
和webapps
至新文件夹catalina-home
下
添加maven配置文件pom.xml
<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
<modelVersion>4.0.0</modelVersion>
<groupId>org.apache.tomcat</groupId>
<artifactId>Tomcat9.0.45</artifactId>
<name>Tomcat9.0.45</name>
<version>9.0.45</version>
<build>
<finalName>Tomcat9.0.45</finalName>
<sourceDirectory>java</sourceDirectory>
<resources>
<resource>
<directory>java</directory>
</resource>
</resources>
<plugins>
<plugin>
<groupId>org.apache.maven.plugins</groupId>
<artifactId>maven-compiler-plugin</artifactId>
<version>2.3</version>
<configuration>
<encoding>UTF-8</encoding>
<source>1.8</source>
<target>1.8</target>
</configuration>
</plugin>
</plugins>
</build>
<dependencies>
<dependency>
<groupId>org.apache.ant</groupId>
<artifactId>ant</artifactId>
<version>1.10.1</version>
</dependency>
<dependency>
<groupId>org.apache.ant</groupId>
<artifactId>ant-apache-log4j</artifactId>
<version>1.9.5</version>
</dependency>
<dependency>
<groupId>org.apache.ant</groupId>
<artifactId>ant-commons-logging</artifactId>
<version>1.9.5</version>
</dependency>
<dependency>
<groupId>javax.xml.rpc</groupId>
<artifactId>javax.xml.rpc-api</artifactId>
<version>1.1</version>
</dependency>
<dependency>
<groupId>wsdl4j</groupId>
<artifactId>wsdl4j</artifactId>
<version>1.6.2</version>
</dependency>
<dependency>
<groupId>org.eclipse.jdt.core.compiler</groupId>
<artifactId>ecj</artifactId>
<version>4.6.1</version>
</dependency>
<dependency>
<groupId>junit</groupId>
<artifactId>junit</artifactId>
<version>4.12</version>
<scope>test</scope>
</dependency>
<dependency>
<groupId>org.easymock</groupId>
<artifactId>easymock</artifactId>
<version>3.5.1</version>
<scope>test</scope>
</dependency>
<dependency>
<groupId>biz.aQute.bnd</groupId>
<artifactId>biz.aQute.bndlib</artifactId>
<version>5.2.0</version>
<scope>provided</scope>
</dependency>
</dependencies>
</project>
./test/trailers
文件夹下写入ResponseTrailers.java
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package trailers;
import java.io.IOException;
import java.io.PrintWriter;
import java.util.HashMap;
import java.util.Map;
import java.util.function.Supplier;
import javax.servlet.ServletException;
import javax.servlet.http.HttpServlet;
import javax.servlet.http.HttpServletRequest;
import javax.servlet.http.HttpServletResponse;
/**
* This example writes some trailer fields to the HTTP response.
*/
public class ResponseTrailers extends HttpServlet {
private static final long serialVersionUID = 1L;
private static final Supplier<Map<String,String>> TRAILER_FIELD_SUPPLIER =
new TrailerFieldSupplier();
@Override
protected void doGet(HttpServletRequest req, HttpServletResponse resp)
throws ServletException, IOException {
resp.setTrailerFields(TRAILER_FIELD_SUPPLIER);
resp.setContentType("text/plain");
resp.setCharacterEncoding("UTF-8");
PrintWriter pw = resp.getWriter();
pw.print("This response should include trailer fields.");
}
private static class TrailerFieldSupplier implements Supplier<Map<String,String>> {
private static final Map<String,String> trailerFields = new HashMap<>();
static {
trailerFields.put("x-trailer-1", "Trailer value one");
trailerFields.put("x-trailer-2", "Trailer value two");
}
@Override
public Map<String, String> get() {
return trailerFields;
}
}
}
./test/util
文件夹下写入CookieFilter.java
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You under the Apache License, Version 2.0
* (the "License"); you may not use this file except in compliance with
* the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package util;
import java.util.Locale;
import java.util.StringTokenizer;
/**
* Processes a cookie header and attempts to obfuscate any cookie values that
* represent session IDs from other web applications. Since session cookie names
* are configurable, as are session ID lengths, this filter is not expected to
* be 100% effective.
*
* It is required that the examples web application is removed in security
* conscious environments as documented in the Security How-To. This filter is
* intended to reduce the impact of failing to follow that advice. A failure by
* this filter to obfuscate a session ID or similar value is not a security
* vulnerability. In such instances the vulnerability is the failure to remove
* the examples web application.
*/
public class CookieFilter {
private static final String OBFUSCATED = "[obfuscated]";
private CookieFilter() {
// Hide default constructor
}
public static String filter(String cookieHeader, String sessionId) {
StringBuilder sb = new StringBuilder(cookieHeader.length());
// Cookie name value pairs are ';' separated.
// Session IDs don't use ; in the value so don't worry about quoted
// values that contain ;
StringTokenizer st = new StringTokenizer(cookieHeader, ";");
boolean first = true;
while (st.hasMoreTokens()) {
if (first) {
first = false;
} else {
sb.append(';');
}
sb.append(filterNameValuePair(st.nextToken(), sessionId));
}
return sb.toString();
}
private static String filterNameValuePair(String input, String sessionId) {
int i = input.indexOf('=');
if (i == -1) {
return input;
}
String name = input.substring(0, i);
String value = input.substring(i + 1, input.length());
return name + "=" + filter(name, value, sessionId);
}
public static String filter(String cookieName, String cookieValue, String sessionId) {
if (cookieName.toLowerCase(Locale.ENGLISH).contains("jsessionid") &&
(sessionId == null || !cookieValue.contains(sessionId))) {
cookieValue = OBFUSCATED;
}
return cookieValue;
}
}
2.2 idea
将项目导入idea,修改Language level
为当前jdk版本,并把catalina-home
文件夹标记为Sources
、test
文件夹标记为Tests
确保jdk版本与环境一致
添加maven国内镜像
<?xml version="1.0" encoding="UTF-8"?>
<settings xmlns="http://maven.apache.org/SETTINGS/1.0.0"
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xsi:schemaLocation="http://maven.apache.org/SETTINGS/1.0.0 http://maven.apache.org/xsd/settings-1.0.0.xsd">
<mirrors>
<!-- 阿里云仓库 -->
<mirror>
<id>alimaven</id>
<mirrorOf>central</mirrorOf>
<name>aliyun maven</name>
<url>http://maven.aliyun.com/nexus/content/repositories/central/</url>
</mirror>
<!-- 中央仓库1 -->
<mirror>
<id>repo1</id>
<mirrorOf>central</mirrorOf>
<name>Human Readable Name for this Mirror.</name>
<url>http://repo1.maven.org/maven2/</url>
</mirror>
<!-- 中央仓库2 -->
<mirror>
<id>repo2</id>
<mirrorOf>central</mirrorOf>
<name>Human Readable Name for this Mirror.</name>
<url>http://repo2.maven.org/maven2/</url>
</mirror>
</mirrors>
</settings>
创建项目执行入口
org.apache.catalina.startup.Bootstrap
-Dcatalina.home=catalina-home
-Dcatalina.base=catalina-home
-Djava.endorsed.dirs=catalina-home/endorsed
-Djava.io.tmpdir=catalina-home/temp
-Djava.util.logging.manager=org.apache.juli.ClassLoaderLogManager
-Djava.util.logging.config.file=catalina-home/conf/logging.properties
修改org/apache/jasper/compiler/JDTCompiler.java
内容,有两处
修改org.apache.catalina.startup.ContextConfig
文件的 configureStart()
方法,初始化 JSP 解析器的代码
context.addServletContainerInitializer(new JasperInitializer(), null);
配置完成,启动项目,访问tomcat
http://127.0.0.1:8080
三、过程分析
3.1 tomct连接器
tomcat中有connector和container两个核心组件,前者对外处理socket连接,后者对内加载和管理servlet,当用户发起请求时,connector收到请求并调用protocolhandler的processor组件解析应用层协议并生成request对象,adaptor再把生成的request对象传递到container做下一步处理
3.2 调试分析
url解析受位于java/org/apache/catalina/connector/CoyoteAdapter
的parsePathParameters()
和normalize()
两个函数影响,打下断点,向/123;456/../index.jsp
路径发起请求
01 parsePathParameters
进入parsePathParameters函数,先判断url中是否含有;
号,不存在时返回-1
并且return
,存在则向下继续执行代码,此时返回分号编号为4
/ 1 2 3 ; 4 5 6 / . . / i n d e x . j s p
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
进入循环,将第二个斜杠后的数据与分号后、斜杠前的pv
数据依次做覆盖操作,然后取前17位数得到处理后的数据uriBC
,函数结束
/ 1 2 3 / . . / i n d e x . j s p . j s p
0 1 2 3 8 9 10 11 12 13 14 15 16 17 18 19 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
02 normalize
进入normalize函数,可以看到对\
、//
、/./
、/../
四种情况进行处理
- 将反斜杠替换为斜杠,
\ ==> /
- 将双斜杠替换为单斜杠,
// ==> /
- 去掉
/./
中的点
- 对
/../
进行跨目录操作
进入循环,每出现一个/../
就向上层目录回溯一次,类似parsePathParameters函数处理分号的逻辑,将/../
之后的数据与/../
之前的数据做覆盖操作,然后取前10位数的到处理后的uriBC
,函数结束
/ 1 2 3 / . . / i n d e x . j s p
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
/ i n d e x . j s p d e x . j s p
1 2 3 4 5 6 7 8 9 10
3.3 小结
先parsePathParameters处理分号,再normalize处理斜杠、反斜杠、点
//aaa;bbb/./../ ==> parsePathParameters() ==> //aaa/./../
//aaa/./../ ==> normalize() ==> /
--> /aaa/./../
--> /aaa/../
--> /
normalize函数对参数进行了一次decodedURI()
操作,所以支持传入url编码后数据,但parsePathParameters函数是直接传入参数,以下地址经过均可访问到主页:
//
/./
/.;/
/aaa/../
/;aaa
/aaa;bbb/../
/aaa;bbb/.././
/aaa/..;./
/aaa;../..;/
/.;/aaa/../
/%2e;/
四、利用场景
4.1 绕过安全设备
某些安全设备存在白名单,对后缀为白名内的请求完全放行,如果服务器使用tomcat作为web容器,则可能存在安全设备绕过风险,如:
- 攻击者向服务器
/sql_vuln/id=1+and+1=1;favicon.ico
发起请求 - 安全设备收到请求,发现到后缀为
ico
,在白名单内不做检测并放行 - 服务器收到请求后对url解析,得到
/sql_vuln/id=1+and+1=1
- 绕过安全设备,成功注入
4.2 绕过访问限制
当后端服务器通过用户当前请求路径来判断权限时,可能存在绕过风险,如:
- 攻击者使用a账户访问admin文件路径,
/;/admin
或/aaa/../admin
- 服务器通过路径来判断当前用户是否具备admin文件访问权限,此时
HttpServletRequest.getRequestURI()
取得的值是原生地址:即/;/admin
,判断不等于/admin
,可以访问 - tomcat对url解析,最终值为
/admin
,攻击者便成功访问/admin
文件
案例:
- Shiro < v1.5.2 身份认证绕过(CVE-2020-1957)
- Shiro < v1.5.3 身份认证绕过(CVE-2020-11989)
- Shiro < v1.6.0 身份认证绕过(CVE-2020-13933)